Setting an Employment Target : Some Considerations Policy Brief Duncan Campbell International Labour Organization Summary The economic crisis of 2008, now having become a major jobs crisis, was one among several factors that is leading a growing number of countries to set explicit employment targets. 1 Such targets can be defined as an explicit political commitment at the highest level to achieve an employment outcome within a specified time period as a principal macroeconomic objective. Often, but not always, the time period for attaining an employment objective falls within a particular political cycle, e.g. an election cycle. As Table 1 below shows, an employment target can be specified in several different ways. Table 1 Examples of employment targets We will cut the unemployment rate by 50% by 2014 (South Africa) We will increase the employment rate to 70% (European Union) We will create 8 million jobs over the next 5 years (Viet Nam) We will guarantee 100 days of work with pay per year per household to anyone who needs it (India) 1 The literature on employment targeting per se is rather scant. There are arguably for relevant strains of literature. The first would be literature examining the role and functioning of planning commissions; the second would be the literature associated with Osami, Islam and others on the growth, employment and poverty nexus ; a third strain, which comes closer to the subject, would be that associated with the University of Massachusetts Political Economy Research Institute, associated with Epstein, Pollin, and Heinz, among others; finally, and closest of all is the work of Per Ronnas on employment constraints and employment targeting, referenced below, and upon which the present paper is partially based. Full references are at the end of this note. 1 P age
Employment targeting is not an alternative to inflation targeting in the sense that it is not a pendular swing away from the importance of macroeconomic stability. That said, there is a difference between the two: the latter assumes implicitly that sound macroeconomic fundamentals are not merely necessary, but sufficient for generating output growth, and, by extension, employment growth; the former assumes that the era of market fundamentalism which has characterized the past quarter century has been insufficient in generating an adequate number of jobs for all those who want to work. Indeed, in the view of many observers, such as the Asian Development Bank, 2 there has been a declining employment intensity of growth, defined as the percentage increase in job growth associated with a 1 % increase in GDP growth. When the highest political authorities make a public commitment to an employment target, the very first question to ask is what is the plan?. In the real world, it could well be that there frankly is no plan. Even without invoking cynicism, it could well be that the announced target is merely a political statement, something that elected officials know will be popular with the electorate. If one did want to invoke a bit of cynicism, there are probably instances in which a promise made in Year 1 could well be forgotten by the time the next election comes around. It is also the case that a government knows that it might not be held to the promise because so many other exogenous events could intervene to frustrate it. And perhaps most cynically of all, a government could conceivably set a target at a low level and be found to have greatly exceeded the target a few years later to the government s political advantage. On the other hand, there could be an ascendency of sophistication with which a target is set. Perhaps most simply, a government could simply have an output growth target in mind. Governments with medium term development plans (such as Nepal) might simply announce a growth target say, an average 5% GDP growth target over the next five years. They know that they grew X% on average over the previous five years, and that Y% employment growth occurred during the period. 3 Based on that relationship the employment elasticity of growth governments could simply use the same ratio, i.e. assume that the employment elasticity will be unchanged in the near future, apply it to forecasted output growth, and derive an employment target. Beyond the rather simple arithmetic above, governments can invoke policies specific to attaining a target. These policies could be horizontal or vertical. A horizontal plan is a policy that applies across the board from which a hoped for employment outcome arises. For example, a general reform improving the time, ease, and cost required to start one s own business. A vertical plan focuses more specifically on sectors. For example, a policy could be designed to promote the garment industry, or to increase jobs and productivity in agriculture. Discussion will return to planning an employment target at a later stage in this brief. 2 3 The ADB s work on this derives from Steven Kapsos work on the employment intensity of growth This appears to be the way that India s Planning Commission approaches the problem. 2 P age
The Policy Challenge While the jobs crisis has focused attention on the centrality of employment, there are underlying factors of longer term. Perhaps most salient among these is the failed assumption, that by assigning a reduced role for the State in the market economy and merely targeting macroeconomic fundamentals, as has been the dominant view since the early 1980s, jobs would appear. The evidence suggested otherwise in many countries that had nonetheless adopted an austere set of macroeconomic targets. The assumption that relatively unfettered market forces would allocate resources including people most efficiently was not borne out in fact. A corollary of this assumption is that policy did not directly target employment, leaving it, rather, as a residual of macroeconomic policy. Many countries that have had a very heady and sustained rate of GDP growth have performed poorly in terms of employment creation. The two largest countries, China and India, fit this picture. In addition to these factors is another: many have pointed to a declining employment intensity of growth as measured by the employment elasticity of output growth. One could assume, of course, that growth today is higher than in the past and is driven by productivity improvements this would in simple arithmetic terms result in an observed decline in employment elasticities. The argument of a declining employment intensity of growth is not predicated on this, however: rather, the argument is made that 4% output growth today is not creating as many jobs as 4% output growth twenty years ago. It is important to note that consensus thus far eludes this matter. The counter argument is that, in the absence of social protection, unemployment is not an option for the vast majority of the world s workforce. People need some sort of income generating activity, and thus are employed doing something. In reconciling the two views, one is on firmer ground in arguing that there has been a decline in the productive employment intensity of growth, as observed in many regions of the world where the growth of the formal economy has stagnated while the informal economy has grown. More jobs may well be needed, but what is certain is that more, better jobs are needed. Policy Options LARGE and ENLARGE Two initial concepts are presented. First, in most countries, the rudimentary data are available to calculate the labour absorbing rate of the growth of employment, abbreviated here as LARGE. This is simply because population data are available, and the people who will enter the labour force in the next five years are alive today. Most simply, the LARGE rate can be calculated under status quo ante assumptions e.g. assuming that there will be no change in the unemployment rate over the next five years, or in the share of the working poor in the labour market. LARGE amounts therefore to new entrants to the labour market, ceteris paribus. An assumption will need to be made about the labour force participation rate (LFPR), the baseline assumption being that it will not change appreciably in the next five years. As a stylized fact, however, we do know that the labour force participation rate of young people is marginally declining, as young 3 P age
people are prolonging their stay in school. An average, recent five year marginal decline in the LFPR for youth would be simple to factor into assumptions of overall LFPR over the ensuing five year period. The concept is a useful one for comparing labour force growth with actual employment growth. Some examples for Asia are shown in Figure 1 below. Figure 1: Labour Force v. Employment Growth, 2000 2006 Employment growth rate (annual, 2000-06, %) 7.0 6.0 5.0 4.0 3.0 2.0 1.0 0.0 Bhutan Afghanistan Philippines Pakistan Mongolia Malaysia Viet Nam Nepal Bangladesh Korea India Thailand Singapore Indonesia Sri Lanka Japan Maldives -1.0-1.0 0.0 1.0 2.0 3.0 4.0 5.0 6.0 7.0 Labour force growth rate (annual, 2000-06, %) 10 In looking at the figure, one sees that Bhutan, while having experienced a high rate of labour force growth, had an even higher rate of employment growth. Whereas, in looking at Nepal, one observes labour force growth that exceeded employment growth over the period. This sort of exercise yields a rudimentary (and inadequate, to which discussion returns) glimpse of labour surpluses and labour shortages. Setting an employment target purely on the basis of the LARGE rate is, however, inadequate. One should be concerned about productive jobs jobs of good quality rather than with just any job. This leads to the second point. One should therefore be interested in the labour absorbing rate of the growth of productive employment, which can be thought of as an enhanced LARGE or, to abbreviate, ENLARGE. This latter rate would include not only new entrants to the labour market, but new entrants into productive jobs, as well as the conversion of existing unproductive jobs into productive ones. Productive employment is the central means for poverty reduction to occur. The Millennium Development Goal poverty target can in this sense be construed as an employment target or more specifically a productive employment target. 4 The MDG poverty target is to halve poverty from its 1990 4 The approach taken in this section uses the methodology developed by Per Ronnas in «A Conceptual and Methodological Guide to Employment Diagnostic Analysis», an unpublished working paper, Geneva, ILO, March 2009. 4 P age
level by 2015. The ENLARGE rate consists of the LARGE rate plus the MDG set target of the reduction of the working poor. In most countries, the productive employment equivalent of reducing poverty can be quantified into a specific productive employment target. Using this rather simple methodology could provide governments with a reality check on their ambitions. For example, in one actual (albeit anonymous) country case, a GDP growth rate of 6.6% through 2015 would be required to achieve the ENLARGE rate, whereas this is more than double recent GDP growth performance in that country. It will be the case in the real world that some employment targets will simply be too ambitious. Attaining an employment target leaves a government with four options: It can recalibrate the target to something more realistic It can evaluate whether a higher rate of GDP growth is feasible It can provide incentives to alter current production technology such that it is more labourabsorbing, 5 or It can through industrial or sectorial policies attempt to change the quality of growth, the pattern of growth. There is an interrelationship, therefore, between employment targeting and industrial policy. While the latter had fallen from favour during the years of the strong policy shift toward free markets, it is currently enjoying a resurgence of interest and for the same reason that employment targeting is; the inadequacy of economic outcomes, including employment outcomes, by leaving everything to markets. The various instruments of industrial policy, from subsidies to taxes to tariffs, among others, form the subject of a separate policy brief. Conclusion ILO interest in employment targeting derives in a very mainstream way from Convention No. 122 on employment policy. The subject has everything to do with the promotion of full, productive and freely chosen employment. As more is knowledge is generated on employment targeting, the ambition will be to have a ready tool for constituents in deriving a credible employment target and technical advice on the means of achieving it. Selected References Islam, R., 2004, The Nexus of Economic Growth, Employment and Poverty Reduction: an Empirical Analysis, Issues in Employment and Poverty Discussion Paper 14, January 2004, Geneva, ILO. 5 Often, seeking to use more labour intensive technology is an unwise policy choice in today s competitive world. There are exceptions, however. Employment intensive infrastructure projects, rather than capitalintensive ones have frequently proven their worth (and is the ILO s second largest technical cooperation activity). It has also been the case that governments have inappropriately subsidized capital at the expense of employment creation 5 P age
Khan, A.R., 2005, Growth, Employment and Poverty: an Analysis of the Vital Nexus based on some Recent UNDP and ILO/SIDA Studies, in Promoting Employment for Poverty Reduction, (Geneva: ILO/UNDP, 2005) Osmani, S.R., 2003, Exploring the Employment Nexus: Topics in Employment and Poverty, Report prepared for the Task Force on the Joint ILO/UNDP Program on Employment and Poverty, Geneva, 2003. Rodrik, D., Normalizing Industrial Policy, Paper prepared for the Commission on Growth and Development, Harvard University, September 2007 Ronnas, P., «A Conceptual and Methodological Guide to Employment Diagnostic Analysis», an unpublished working paper, Geneva, ILO, March 2009. www.peri/umass.edu 6 P age